No more OP death stacks

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
wobbly
Cosmic Dragon
Posts: 1989
Joined: Thu Oct 10, 2013 6:48 pm

Re: No more OP death stacks

#46 Post by wobbly »

Ophiuchus wrote: Sat Apr 06, 2024 8:09 am
wobbly wrote: Sat Apr 06, 2024 6:45 am
Ophiuchus wrote: Sat Mar 30, 2024 1:53 pm

The main downside of this is that it gives an incentive to separate all your ships in times of peace and to amass them if going for battle.
Charge the cost to a "command vessel", that way the asset has to exist in advance.
I don't understand how you want this to work. What about ships in a system which go over the stack limit/without command vessel?
You mean no fleet without command vessel is allowed to actively participate in combat or something like this?
So I was deliberately vague, because the exact implementation depends on what you are trying to achieve, so you'd fit the mechanics to the result. The most basic ways of implement command/logistics/support vessels is to either give a buff if you have sufficient command for the number of ships or a penalty for insufficient ships. If that's not enough you could restrict the number of ships taking place in combat (and also consider whether excess ships can join in later rounds if ships are destroyed) or some other solution.

The main point of the idea is to charge IP upkeep to the command vessel instead of the stack, so that moving ships in and out of stacks doesn't effect the upkeep. Then off course you need some kind of mechanic that makes big stacks ineffective without sufficient command vessels present.

Not saying its necessarily a good solution, its just one way to make the IP upkeep more constant, rather then constantly changing every time a ship moves. Charging IP dependent on where a ship is sounds like a nightmare to me. The IP cost is going to constantly change, that's a management headache. How much IP do you need? You suddenly don't know.

Note: Any solution has to also work for the AI. Teaching it how to layout its ships to handle IP costs effectively isn't going to be simple.

Ophiuchus
Programmer
Posts: 3484
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: No more OP death stacks

#47 Post by Ophiuchus »

wobbly wrote: Mon Apr 08, 2024 6:26 am Charging IP dependent on where a ship is sounds like a nightmare to me. The IP cost is going to constantly change, that's a management headache. How much IP do you need? You suddenly don't know.

Note: Any solution has to also work for the AI. Teaching it how to layout its ships to handle IP costs effectively isn't going to be simple.
ok, these are very valid points.

could you please comment on my ideas on command vessels so i see if/where i am talking bullshit?
Oberlus wrote: Sun Apr 07, 2024 11:17 pm Then make the upkeep of a given ship dependent on the count of non-enemy ships in the vicinity
My suggestion is to count visible peaceful & allied ships although it is not 100% correct. Is that different from your suggestion?

I am sure it is not OK to include invisible ships. I am not sure if it is OK to only include visible ships. The best option upkeep-wise for my allies and me would be to change to peaceful and make the ships stealthy so the "allies" are not able to detect those and split the combat stack in ~equal parts.

Actually on fine print for the Oberlus(-inspired) design (my preference in bold(:
  • there are different options on which parameters a single ship in vicinity could be weighted for upkeep amount. I think using ship count (or ship part count) like we currently have for PP upkeep is probably fine; Note this is way better for big ships. But there would be also options to base it on PP cost of ship, PP cost of hulls, combat power or similar.
  • there are different options to measure the distance/vicinity. In one dimension taking uu-distance along starlanes, hops along starlanes, or direct uu-distance. If not using hops, there is the choice of taking the ship speed meter or simply a fixed uu distance.
  • I would prefer to consider visible non-enemy fleets in a vicinity of either (e.g. 2) starlane hops or basing it on a fixed uu distance (e.g. 160uu) vicinity.
  • I would base the distance calculations on next destination systems rather than on real fleet positions on starlanes in order to take away incentives for decreasing upkeep by having the fleets on starlanes. This would (with reification) also drastically decrease the necessary computation; a mostly constant map<system, list<systems>> would suffice; recomputation would only be necessary if distances or topology of systems changes. Note this may be necessary for hops/distance between systems; but might not be necessary if measuring using uu distance between fleets.
  • make the formula (e.g. influence_upkeep_per_turn(ships-count) = (positive_or_zero(ships-count) - 10)^2 or influence_upkeep_per_turn(sum-of-ship-cost-in-PP) = positive_or_zero(sum-of-ship-cost-in-PP - 500)^2 ) square-based , so one has to pay linear to the theoretical amount of gain in the combat power formula. Note the gain for small ships is lower than for big ships (because of attrition and the difference between max-hits-to-live and structure).
  • having a cutoff in the formula (i.e. just charge after a certain threshold is reached) is good for only introducing this complexity at a later point. Note this means the cost will be truly square based only for infinitely big fleets.
  • on the design page add a line how much a certain fleet size of this design is gonna cost upkeep. Like: having 50 ships of this design in a vicinity will cost xx IP/per turn upkeep (link to pedia-IP-upkeep-concept)
  • if not basing the amount on count: on the design page add a line how how big a fleet can get without paying upkeep like: you can have 10 ships of this design in a vicinity without having to pay IP upkeep (link to pedia-IP-upkeep-concept)
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
LienRag
Cosmic Dragon
Posts: 2241
Joined: Fri May 17, 2019 5:03 pm

Re: No more OP death stacks

#48 Post by LienRag »

Oberlus wrote: Sun Apr 07, 2024 11:17 pm Then make the upkeep of a given ship dependent on the count of non-enemy ships in the vicinity (maybe half the weight of peaceful ships).

Oh, vicious !
A way to cripple the Influence of your frenemies !
I like it !
Also, it makes the "blockade" stance much more relevant...

User avatar
Vezzra
Release Manager, Design
Posts: 6114
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: No more OP death stacks

#49 Post by Vezzra »

Ophiuchus wrote: Fri Apr 05, 2024 1:07 pm
o01eg wrote: Fri Apr 05, 2024 4:24 am What if make it possible for weapon to miss and reduce chances to missing with bigger enemy concentration?
this is not in the design space until geoff and vezzra shift their position
I'd be very cautious changing this.

Having weapons always hit and not including hit/miss chances has been part of the original core design and design philosohpy, and is there for a good reason. IIRC the consensus has been that weapons having a hit/miss chance was adding complexity, but not adding fun/interesting dynamics/choices which would make that added complexity worth it.

Also, a lot of stuff (techs, weapons, armor, shields and ship stats, combat mechanics etc.) has been designed and balanced based on that design decision. If you change that, you'll have to redesign and rebalance a lot of other stuff too.

In this particular case, if the idea is to introduce a disadvantage of having too many ships in the same space because more ships per amount of space means it's easier to hit them in combat, so giving the party with the fewer ships an edgde, maybe this could be achieved by increasing damage? Thus kind of simulating that "more" hits "hit home", or explosions caused by hits cause damage to nearby ships caught in the blast radius or something along these lines.

Just an idea.

User avatar
LienRag
Cosmic Dragon
Posts: 2241
Joined: Fri May 17, 2019 5:03 pm

Re: No more OP death stacks

#50 Post by LienRag »

Actually having Arc Disruptors do more damage against huge stacks (and not only get more shots) would probably be fine.
It would of course make it a weapon of choice even more than now, but since it's easily countered by shields that's not much of a problem (and on the opposite, would resurrect shields).

wobbly
Cosmic Dragon
Posts: 1989
Joined: Thu Oct 10, 2013 6:48 pm

Re: No more OP death stacks

#51 Post by wobbly »

LienRag wrote: Fri Apr 12, 2024 12:33 pm It would of course make it a weapon of choice even more than now
Is this a statistically provable fact, or just some sort of carte blanche statement held up by nothing?

o01eg
Programmer
Posts: 2045
Joined: Sat Dec 10, 2011 5:46 am

Re: No more OP death stacks

#52 Post by o01eg »

Vezzra wrote: Fri Apr 12, 2024 10:54 am Thus kind of simulating that "more" hits "hit home", or explosions caused by hits cause damage to nearby ships caught in the blast radius or something along these lines.

Just an idea.
That reminds me somewhat of stack kill in FreeCiv.
Gentoo Linux x64, gcc-11.2, boost-1.78.0
Ubuntu Server 22.04 x64, gcc-12, boost-1.74.0
Welcome to the slow multiplayer game at freeorion-lt.dedyn.io.Version 2024-05-12.bb52512.
Donations're welcome:BTC:bc1q007qldm6eppqcukewtfkfcj0naut9njj7audnm

Ophiuchus
Programmer
Posts: 3484
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: No more OP death stacks

#53 Post by Ophiuchus »

Vezzra wrote: Fri Apr 12, 2024 10:54 am Thus kind of simulating that "more" hits "hit home", or explosions caused by hits cause damage to nearby ships caught in the blast radius or something along these lines.
simulating chance-to-(meaningful)-hit by amount-of-damage is definitely possible with the current backend etc.
To get something like this right (changing damage based on the amount of enemies), we would need pre-combat effects though. Having a 1-turn delay is probably too late (because the ships which could benefit probably do not exist anymore) or a bit weird (basing it on known potential enemies in some distance).

if we had such a reasoning, it might also make sense to add shields by default and add more shields for "nimble" ships (good at "evasive maneuvers").
This would make more sense with 1-turn delay than the other way round. On one hand because on the stacking side the own ships are always known.

Doing both (adding damage if there are lots of enemies and decreasing shields when there are a lot of allies/peacies) would partially address visibility/relationship concerns.

Doing it rights would use the real numbers (without visibility), but it would needs to happen only if combat happens, directly before combat (or in-combat) in order not to leak too much hidden info and to know the real situation.

Hm. Actually this feels a lot like my original suggestion. Just more discrete and ad hoc (some weapons having certain boni when there are certain amounts of enemies).
o01eg wrote: Sat Apr 13, 2024 5:10 am That reminds me somewhat of stack kill in FreeCiv.
I am not familiar with it. Found this "Stack kill
In titles prior to Civilization III, a combat mechanic by which defeating the defending unit eliminates all other units in the stack."
Could you elaborate?
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

o01eg
Programmer
Posts: 2045
Joined: Sat Dec 10, 2011 5:46 am

Re: No more OP death stacks

#54 Post by o01eg »

Ophiuchus wrote: Sun Apr 14, 2024 7:34 pm
o01eg wrote: Sat Apr 13, 2024 5:10 am That reminds me somewhat of stack kill in FreeCiv.
I am not familiar with it. Found this "Stack kill
In titles prior to Civilization III, a combat mechanic by which defeating the defending unit eliminates all other units in the stack."
Could you elaborate?
I mean in FreeCiv stacks issue mitigated by killing entire stack if the stack's defender gets killed. I suppose additional damage to stack's ships if something gets destroyed is the somewhat same approach.
Gentoo Linux x64, gcc-11.2, boost-1.78.0
Ubuntu Server 22.04 x64, gcc-12, boost-1.74.0
Welcome to the slow multiplayer game at freeorion-lt.dedyn.io.Version 2024-05-12.bb52512.
Donations're welcome:BTC:bc1q007qldm6eppqcukewtfkfcj0naut9njj7audnm

User avatar
Oberlus
Cosmic Dragon
Posts: 5789
Joined: Mon Apr 10, 2017 4:25 pm

Re: No more OP death stacks

#55 Post by Oberlus »

o01eg wrote: Mon Apr 15, 2024 6:31 am
Ophiuchus wrote: Sun Apr 14, 2024 7:34 pm Could you elaborate?
I mean in FreeCiv stacks issue mitigated by killing entire stack if the stack's defender gets killed. I suppose additional damage to stack's ships if something gets destroyed is the somewhat same approach.
What I remember from last time I played it, a few years back, is that killing a unit in a stack makes all units in that stack get damaged with a sort of AOE. So by killing a single unit would not kill the whole stack, but killing a few would.

We can mimic that effect with Vezzra's suggestion, giving a bonus to damage against bigger fleets, so that less shots are needed to kill a single ship and hence more ships are killed.
I would make it happen in the backend to work around the 1-turn delay issue.
But how to inform the players?!
The fleet/ship damage figures in the UI before the combat are a complex problem that I'd like to just ignore: When fleet A is at system A, next to fleet B at system B, what damage should it show to player A and player B depending on their fleet sizes? And should it take into account current movement orders of your own fleet and known orders of foreign fleets? And should it take into account the reinforcements that will be there next turn? And what if there is another fleet C at system C, that is allied to one of A or B, or not?
Quite the shitshow.

The work around this UI problem could be to use an easy rule? Something like a multiplier based on the ratio between own and enemy fleet size, like "my ships' damage is multiplied by sqrt(their_ships/my_ships)", so if I have 4 times more ships my lasers will do half the damage. And leave damage of the smaller fleet untouched.
Oh, that could make shielded ships in small fleets against big ships immune to gun shots...

OK, what about reducing number of shots instead of changing damage?
Edit: but not as in that suggestion about missing shots (chance to miss), but with a deterministic equation, so not a chance, a fixed ratio.

User avatar
Grummel7
Space Dragon
Posts: 341
Joined: Mon Oct 09, 2017 3:44 pm

Re: No more OP death stacks

#56 Post by Grummel7 »

o01eg wrote: Mon Apr 15, 2024 6:31 am I mean in FreeCiv stacks issue mitigated by killing entire stack if the stack's defender gets killed. I suppose additional damage to stack's ships if something gets destroyed is the somewhat same approach.
The entire stack is only destroyed if the defender is not in a city or fortress. Maybe some defender damage was added at some point because walled cities are notoriously hard to conquered until you got artillery. But the mechanics of civilization/freeciv don't fit for a space game anyway. In civ the attacker chooses one unit, which fights one unit of the defender, etc. Which may kind of resemble ancient war fare, but would be totally weird for space battles.

The big problem we have is that in a scenario where one player has 20 ships and another has 10, a typical outcome is 20 slightly damaged ships on one side and 10 destroyed on the other. Considering the ships are operated by intelligent beings, the 10 would either try to flee or if this is not an option, concentrate their firepower on a few ships to at least do some lasting damage.

I wonder if we could add a formation to fleets. In films you usually see battle fleets as big clusters of ships but the fleet may as well spread randomly over the entire system and moving around quickly. If using such a formation, they will have reduced firepower, but the opponents may also miss some / most of their shots (chance depending on stealth vs. detection).

User avatar
Vezzra
Release Manager, Design
Posts: 6114
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: No more OP death stacks

#57 Post by Vezzra »

Oberlus wrote: Tue Apr 16, 2024 11:54 amThe fleet/ship damage figures in the UI before the combat are a complex problem that I'd like to just ignore: When fleet A is at system A, next to fleet B at system B, what damage should it show to player A and player B depending on their fleet sizes? And should it take into account current movement orders of your own fleet and known orders of foreign fleets? And should it take into account the reinforcements that will be there next turn? And what if there is another fleet C at system C, that is allied to one of A or B, or not?
Quite the shitshow.
The basic issue we're facing here is a Can't-eat-the-cake-and-keep-it thing: Deep and interesting combat mechanics are by necessity more complex, which makes it hard (if not impossible) to quantify the combat strength of a fleet. Because that essentially is the point: You don't want a combat system where you can look at a simple number that tells the strength of fleets and by that how they will fare in combat against each other. You want a whole set of parameters/stats that require you to think and plan, because the specific combination of all the stats of all involved fleets determine how combat will play out.

If however you want to be able to easily determine a fleets combat strength compared to other fleets, you need a sufficiently simple and straightforward combat system.

So, we need to decide what we want from space combat in FO: deep or simple. Both approaches have pros and cons, but we can only have one, not both. Space combat is either deep or simple and straightforward, but we can't max out both. We need to find a compromise. The more depth we want, the harder it will get to quantify fleet combat strength by some number, and vice versa.

User avatar
Vezzra
Release Manager, Design
Posts: 6114
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: No more OP death stacks

#58 Post by Vezzra »

Grummel7 wrote: Wed Apr 17, 2024 7:41 pmI wonder if we could add a formation to fleets. In films you usually see battle fleets as big clusters of ships but the fleet may as well spread randomly over the entire system and moving around quickly. If using such a formation, they will have reduced firepower, but the opponents may also miss some / most of their shots (chance depending on stealth vs. detection).
My solution to this challenge would be to redesign combat more fundamentally. As you point out, space combat in FO isn't necessarily the classic fleet battle scene in scify, where big clusters of ships face off against each other. It probably should be actually seen as a series of battles/skirmishes that take place all over the star system during a single game turn.

I'd split each combat bout into two phases: maneuver and firing. The maneuvering phase would model the participating fleets trying to outmaneuver each other, each proponent trying to get their forces into positions where they can deal the most damage to the enemy while minimizing the damage inflicted on themselves. By that I mean trying to move their forces in a way that in the resulting actual battles/skirmishes their own forces are matched as advantageous against the enemy forces as possible.

Technically that would mean for combat resolution, that during this phase the algorithm determines a series of engagements that will happen during the current combat bout. Then, for each engagement determine which ships on each side will participate in that particular engagement. Each ship can only participate in not more than one engagement. The better one side maneuvers their forces compared to the other side, the more of their total ships they will be able to bring to the engagements, and the more advantageous the resulting pairings will be for them.

This way, a fleet that is outnumbered/outgunned, can potentially counter that by being superior at maneuvering.

The maneuvering capabilities of ships/fleets would be determined by ship stats (stealth and detection would play a major part here!), big, slow ships being at a disadvantage here to small, nimble ships, by techs, policies, species traits etc. If we implement leaders at some point, fleet admirals could play a major part.

During the firing phase combat resolution would work the same as now for a single combat bout. But not, as it is now, with all the participating ships in the entire system facing off against each other in one big engagement. Instead, you'd have several engagements (as determined in the maneuvering phase) with only the ships participating in the same engagement involved.

Paradox Victoria 3 front line war model works basically this way (or very similar). Given the time scale of a turn in FO, what happens when space combat occurs in a star system isn't really a single, massive battle, but actually more of an entire war consisting of a series of battles. Treating it as such and modelling it accordingly makes sense IMO.

Just an idea/suggestion. But I think it might help to solve quite some balancing issues and would give us effective means especially to mitigate the issue that the side with the firepower advantage usually gets away with minimal losses, while the inferior side gets completely wiped out.

Ophiuchus
Programmer
Posts: 3484
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: No more OP death stacks

#59 Post by Ophiuchus »

Vezzra wrote: Wed May 01, 2024 10:40 am
Grummel7 wrote: Wed Apr 17, 2024 7:41 pmI wonder if we could add a formation to fleets. In films you usually see battle fleets as big clusters of ships but the fleet may as well spread randomly over the entire system and moving around quickly. If using such a formation, they will have reduced firepower, but the opponents may also miss some / most of their shots (chance depending on stealth vs. detection).
Just an idea/suggestion. But I think it might help to solve quite some balancing issues and would give us effective means especially to mitigate the issue that the side with the firepower advantage usually gets away with minimal losses, while the inferior side gets completely wiped out.
i think both are out of scope for this thread (would need its own topic); here I am looking for a (maybe suboptimal) way to solve the problem (which you very well stated) in the bounds of the current combat system.

One thing which is connected to this thread though: currently we get a clear winner for the system pretty fast

If it is better possible for a part of the loosing fleet to survive a battle we probably need to adapt blockade rules if the loosing fleet was doing the blockade. I think it would be very frustrating to be hold up continuously by a comparatively small force. (side note: if only ships actually being hit in the battle can be effectively blockaded, this would not be problem)

I think this could be relevant for some solutions inside of the scope of the thread. Note this is not a problem for the oberlus-inspired design, tweaking not the battle but the fleet upkeep. The original design (suggesting a scaling of combat power) would declare ships as conquered if being reduced structure "killed" and would not count conquered ships for keeping up a blockade.

Are there any objections to trying the oberlus-inspired design?
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
LienRag
Cosmic Dragon
Posts: 2241
Joined: Fri May 17, 2019 5:03 pm

Re: No more OP death stacks

#60 Post by LienRag »

Ophiuchus wrote: Wed May 01, 2024 11:10 am (side note: if only ships actually being hit in the battle can be effectively blockaded, this would not be problem)

That may be interesting indeed.




Ophiuchus wrote: Wed May 01, 2024 11:10 am Are there any objections to trying the oberlus-inspired design?

Well, maybe first try the LienRag-inspired design ?
(not the total rework that I advocated for years ago and am still not able to propose a PoC for, just the new weapons with damage depending of number of ships in the opposing fleet)
It's simple and incremental to what we have, I believe that it's better to first check whether a small change like that could help fix the balance.

Post Reply